policy

canvas_agentic_gym

avasile96 · PyTorch

or hover any field below to flag it

Overview

Name
canvas_agentic_gym
Author
avasile96
Framework
PyTorch
License
unknown
Skill type
manipulation
Evidence level
untested
Task description
A minimalist 2D canvas that exposes a standard RL interface (State, Action, Reward) and simultaneously acts as an MCP server for LLM tool-calling. No model training required -- focus is on environment design, MDP formulation, and API tooling.

Spaces

Action space
other · 0-dim · 0Hz
Observation space
  • type: other

Links

HuggingFace repo
null
Paper (arXiv)
null

Compatible robots

20

Compatible environments

0

No environments list canvas_agentic_gym yet.

Datasets that reference this policy

0

No datasets reference canvas_agentic_gym yet.